Involvement of Active Oxygen Species in Degradation of the D1

An Identifiable State Model To Describe Light Intensity Influence on ... Hydrogen Peroxide in Darkness: Implications for the Mechanism of Degradation ...
0 downloads 0 Views 3MB Size
9722

Biochemistry 1994, 33, 9722-9730

Involvement of Active Oxygen Species in Degradation of the D1 Protein under Strong Illumination in Isolated Subcomplexes of Photosystem IIt Mitsue Miyao Laboratory of Photosynthesis, National Institute of Agrobiological Resources (NIAR),Kannondai, Tsukuba 305, Japan Received January 19, 1994; Revised Manuscript Received June I, 1994'

I1 (PSII) were investigated using three different isolated subcomplexes of PSII, namely, the PSII complex depleted of major lightharvesting proteins, the core complex, and the reaction center complex. Under illumination, not only the D1 protein of the reaction center but also other intrinsic proteins sustained some damage in all three subcomplexes: Coomassie blue-stained bands after polyacrylamide gel electrophoresis were smeared, and their migration distances on the gel were reduced with increasing duration of illumination. Such damage occurred first in the D1 and D2 proteins and subsequently in the 43- and 47-kDa proteins of the core antenna and the subunit of cytochrome bss.+ Immunoblot analysis using an antibody specific to the D1 protein showed that the D1 protein was degraded to major fragments of about 23 and 16 kDa during illumination. The smearing and changes in mobility of protein bands, as well as the fragmentation of the D1 protein, were greatly suppressed by scavengers of active oxygen species. From the effectiveness of scavengers, it appeared that superoxide anions participate in the protein damage in the PSII complex, hydrogen peroxide in the PSII and core complexes, and singlet oxygen, hydroxyl, and alkoxy1radicals in all three subcomplexes. We also found that fragments of the D1 protein of 23 and 16 kDa were formed even when PSII complexes that had been completely solubilized with sodium dodecyl sulfate were illuminated. This fragmentation was also suppressed by active oxygen scavengers. These observations suggest that in isolated PSII subcomplexes under strong illumination the D1 protein is cleaved at specific sites solely by the action of active oxygen, and that the D1 protein has amino acid sequences specifically susceptible to attack by active oxygen. ABSTRACT: The effects of strong illumination on the proteins in photosystem

Strong illumination of oxygenic photosynthetic organisms results in loss of their photosyntheticcapacity (Powles, 1984). The primary event in this process is impairment of photosystem I1 (PSII;' Kyle, 1987). The impairment of PSII during illumination (photoinhibition) involves photoinactivation of PSII activity and specific degradation of the D1 protein of the photochemical reaction center (Prhiil et al., 1992; Aro et al., 1993). The photoinhibition is considered to proceed by two different mechanisms, so-called acceptor-side and donor-side photoinhibition (Barber &Anderson, 1992;Aroet al., 1993). The acceptor-sidephotoinhibition occurs in materials in which the donor side of PSII is functional. Illumination initially affects the acceptor side of PSII and blocks the electron flow while the primary photochemical process continues to occur (Vass et al., 1992). Under these conditions, the triplet state of P680 is formed with high probability and generates toxic singlet oxygen ('02) in a reaction with oxygen. Donor-side photoinhibition is observed in materials in which the donor side is inactivated. The initial event is impairment of electron t This work was supported in part by a Grant-in-Aid for Cooperative Research (04304004) from the Japanese Ministry of Education, Science and Culture. @Abstractpublished in Advance ACS Abstracts, July 15, 1994. Abbreviations: anti-D1, antibody raised against the D1 protein; APMSF, (4-amidinopheny1)methanesulfonyl fluoride;Chl, chlorophyll; Cyt, cytochrome;DABCO, 1,4-diazabicyclo[2.2.2]octane;DBMIB, 2,sdibromo-3-methyl-6-isopropyl1,4-benzoquinone;DFP, diisopropyl fluorophosphate; DM, n-dodecyl 8-Dmaltoside; Hepes, N(2-hydroxyethy1)piperazine-N'-2-ethanesulfonicacid;LHCII, light-harvestingchlorophyll complex of photosystem II; Mes, 2-(N-morpholino)ethanesulfonic acid; Mops, 3-(N-morpholino)propanesulfonic acid; P680, primary electron donor of photosystem 11; PAGE, polyacrylamide gel electrophoresis; PMSF, phenylmethanesulfonyl fluoride; PSII, photosystem 11; RC, reaction center; SBTI, soybean trypsin inhibitor; SDS, sodium dodecyl sulfate; SOD, superoxide dismutase; TPCK, N@-tosyl-L-phenylalanine chloromethyl ketone; Tris, tris(hydroxymethy1)aminomethane.

0006-2960/94/0433-9722%04.50/0

transfer from Tyrz, the secondary electron donor to PSII, to P680+ (Blubaugh et al., 1991; Eckert et al., 1991), which leads to stabilization of strongly oxidizing species, such as P680+, Chl+, and Tyr+. '02 and the oxidizing species generated in this way cause irreversible damage to the reaction center and induce the subsequent degradation of the D1 protein. The D1 protein is known as a rapid turnover protein which is specifically degraded under illumination in vivo (Mattoo et al., 1981, 1984). The degradation in vivo gives rise to a fragment of 23.5 kDa (Greenberg et al., 1987). Proteolytic mapping of the fragment (Greenberg et al., 1987) revealed that the cleavage site is located in the loop that connects the membrane-spanning helixes IV and V of the folding model of the D1 protein proposed by Trebst (1986). Subsequently, a number of studies were performed in vitro, and it was demonstrated that D1 protein degradation occurs even in PSII membranes and PSII subcomplexesunder strong illumination (Aro et al., 1993). In PSII subcomplexes,various fragments of the D1 protein can be observed after strong illumination, and a fragment of 23-24 kDa is considered to be the primary degradation product (Aro et al., 1993). The primary cleavage site varies depending on the illumination conditions. When oxygen-evolving subcomplexes are illuminated under conditions that support oxygen evolution (acceptor-side photoinhibition), the D1 protein is cleaved in the loop connecting helixes IV and V on the stromal side of the thylakoid membrane, giving rise to a 23-kDa fragment of N-terminal origin, as occurs in vivo (Salter et al., 1992; De Las Rivas et al., 1992). Such cleavage also occurs when isolated RC complex, which lacks oxygen-evolvingcapacity, is illuminated in the absence of electron acceptors (De Las Rivas et al., 1993). When illuminated in the presence of DBMIB as an acceptor under conditions that do not permit oxygen evolution 0 1994 American Chemical Society

D1 Protein Degradation and Active Oxygen (donor-side photoinhibition), cleavage occurs in the loop connecting helixes I and I1 on the lumenal side of the thylakoid membrane, giving rise to a 24-kDa fragment of C-terminal origin (De Las Rivas et al., 1992). With respect to the mechanism of degradation of the D1 protein, involvement of specific proteases has long been implicated. Recent in vitro studies suggest that serine-type protease(s) intrinsically present in PSII, possibly a component of PSII itself, catalyze(s) the degradation (Barber & Andersson, 1992; Aro et al., 1993). This possibility is inferred from the observation that inhibitors of serine-type proteases suppressed D1 protein degradation in isolated PSII subcomplexes. According to this model, ' 0 2 or oxidizing species generated inside the reaction center alter the conformation of the D1 protein and render it susceptible to degradation by the putative protease(s). Some uncertainty exists in this protease model. The identity of the protease is controversial. From the binding characteristics of DFP, a covalent blocker of the catalytic site of serine-type proteases, Salter et al. (1992) proposed that the 43-kDa protein of the core antenna might be the protease. On the other hand, Barber and co-workers proposed that one of the reaction center components (D1 and D2 proteins, Cyt b559, and the product of psbl) has proteolytic activity, since the inhibitors were effective even in isolated RC complexes (Shipton & Barber, 1992; De Las Rivas et al., 1993). The number of putative proteases is also open to question. There are at least two primary cleavage sites of the D1 protein, and they are located on opposite sides of the thylakoid membrane. In addition, another primary cleavage site which gives rise to a fragment of 16 kDa has been proposed (Barbato et al., 1992a). Even if only the primarycleavages are each catalyzed by a protease, there should be at least three different proteolytic activities in PSII. Another mechanism for D1 protein degradation has been proposed, namely, involvement of active oxygen species generated in PSII under illumination. Bradley et al. (1991) proposed that hydrogen peroxide (H202) generated at the specific sites of PSII could cause the degradation. Kyle (1987) proposed that a plastosemiquinone radical in the QB-binding site would generate superoxide anion (02-) and hydroxyl radical (.OH) in a reaction with oxygen and these damage the D1 protein. Sopory et al. (1990) investigated the effects of scavengers in vivo and suggested that not ' 0 2 but some other oxygen radical is involved in the degradation. Similarly, from the effectivenessof scavengers and antioxidants, various active oxygen species have been proposed to participate in the photoinactivation of PSII and D1 protein degradation in isolated thylakoids (Barhyi & Krause, 1985; Richter et al., 1990; Tschiersch & Ohmann, 1993) and PSII membranes (Setllk et al., 1990; Chen et al., 1992). In the present study, the possible involvement of active oxygen species in degradation of the D1 protein was investigated with three different PSII subcomplexes, namely, the PSII complex depleted of the major LHCII, the core complex, and the RC complex. We found that, in these isolated subcomplexes, various different species of active oxygen are involved in D1 protein degradation, and that the D1 protein can be cleaved at specific sites solely by the action of active oxygen.

MATERIALS AND METHODS Preparation of PSII Membranes and Isolation of PSII Subcomplexes. PSII membranes were prepared from 3-5week-old rice seedlings with Triton X-100 by the method for wheat (Miyao & Inoue, 1991) with slight modifications, and

Biochemistry, Vol. 33, No. 32, 1994 9723 stored in liquid nitrogen in the presence of 30% (v/v) ethylene glycol. Before use, the membranes were thawed and washed 3 times with 10 mM NaC1,0.4 M sucrose, and 25 mM MesNaOH (pH 6.5; medium A) by centrifugation and resuspension. The oxygen-evolving activity of the PSII membranes was around 700 pmol (mg of Chi)-' h-' with 0.8 mM phenyl1,4-benzoquinone as an electron acceptor. PSII complexes were prepared from PSII membranes as described by Kashino et al. (1992) with some modifications as follows. The PSII membranes were pelleted by centrifugation at 40000g for 10 min and suspended in 20 mM NaHCO3, 6.25 mM CaC12, 0.5 M NaCl, 1.25 M sucrose, and 20 mM Hepes-NaOH (pH 7.0) at 3 mg of Chl/mL. The suspension was supplemented with 1/100 volume of 4.0% (w/v) Triton X-100, mixed thoroughly, and allowed to stand on ice for 10 min. This mixture was supplemented with 1/4 volume of 10% (w/v) n-heptyl 8-D-thioglucoside to give a final detergent concentration of 2%. The mixture was mixed thoroughly, allowed to stand on ice for 15 min, and then diluted with 2 volumes of 20 mM NaHC03, 5 mM CaC12, 0.4 M sucrose, and 20 mM Hepes-NaOH (pH 7.0). After standing on ice for 5 min, the mixture was centrifuged at 35000g for 15 min, and the supernatant at 40000g for 10 min. The resultant supernatant was dialyzed against 20 mM NaHCO3, 10 mM NaCl, 0.4 M sucrose, and 40 mM Mes-NaOH (pH 6.0) for 3 h. The dialyzate was centrifuged at 35000g for 15 min, and the resultant pellet consisting of the PSII complexes was suspended in medium A. Core complexes were prepared by treatment of the PSII membranes with n-octyl 0-D-glucopyranoside followed by sucrose density gradient centrifugation according to Ikeuchi and Inoue (1988) and finally suspended in medium A. The oxygen-evolving activity of the PSII complexes was 2000 pmol (mg of Chi)-' h-I with 0.8 mM phenyl- 1,4-benzoquinone, and that of the core complexes was 1800 pmol (mg of Chi)-' h-I with 0.8 mM 2,6-dichloro-1,4benzoquinone. RC complexes were prepared from PSII membranes with Triton X-100 by a method based on that of Nanba and Satoh (1987) as modified by Chapman et al. (1988), with the exception that alkaline treatment of the PSII membranes prior to the Triton treatment was omitted and the second chromatographic separation was performed in the presence of 2 mM DM instead of 0.2% Triton X-100. The photochemical activity of the RC complexes, measured at 25 OC as the reduction of 0.25 mM silicomolybdate with 1 mM MnClz as an electron donor (Shipton & Barber, 1992), was 200-300 pmol (mg of Chl)-' h-I. The isolated PSII subcomplexes were frozen in liquid nitrogen and kept at -80 OC until use. All procedures were performed under dim light at 0-4 OC. Chl was determined according to Arnon (1949). Photoinhibitory Light Treatments. The PSII and core complexes were subjected to acceptor-side photoinhibition essentially as described previously (Virgin et al., 1991; De Las Rivas et al., 1992), and the RC complexes were subjected to either acceptor-side (De Las Rivas et al., 1993) or donorside (Shipton & Barber, 1992) photoinhibition. The PSII and core complexes were suspended in 1 mM DM, 10 mM NaC1, 0.4 M sucrose, and 50 mM Mes-NaOH (pH 6.0) at 100 pg of Chl/mL. The RC complexes (1 50-200 pg of Chl/ mL, 110 mM NaCl, pH 7.2) were diluted to 50 pg of Chl/mL with 2 mM DM and 0.4 M sucrose containing either 50 mM Mes-NaOH (pH 6.0) or 50 mM Tris-HC1 (pH 8.5) to give a final pH of 6.3 or 8.0. After standing in darkness at 20 OC for 10min, the suspension was illuminated in a glass cuvette,

9724 Biochemistry, Vol. 33, No. 32, 1994

A

47

1

D2 -

C

B

0 1020304560 0 min

33Ex

Miyao

0 1020304560 Omin

D

0 1 2 4 8 0 min

0 5 102040600 min

Y

5

b 559 II

-

B

23

I

-23

24

-16

16

16 10

FIGURE1: Changes in polypeptide and immunoblot profiles of PSII subcomplexes during photoinhibitory illumination. The PSII and core complexes were suspended in 1 mM DM, 10 mM NaCl, 0.4 M sucrose, and 50 mM Mes-NaOH (pH 6.0) at 100 pg of Chl/mL, and the RC complexes were suspended in 2 mM DM and 0.25 M sucrose containing 50 mM Mes-NaOH (pH 6.3) or 50 mM Tris-HC1 (pH 8.0) at 50 pg of Chl/mL. After standing in darkness at 20 "C for 10 min, the suspension was illuminated with white light for the designated times in the presence or absence of 0.2 mM DBMIB at 20 OC. (A) PSII complexes illuminated at 8.0 mE/(m2*s). (B) Core complexes illuminated at 8.0 mE/(m2*s). (C) RC complexes illuminated at 2.0 mE/(m2*s) and pH 6.3. (D) RC complexes illuminated at 4.0 mE/(m2-s) and pH 8.0 in the presence of DBMIB. (Upper panels) Polypeptide profiles after staining with Coomassie blue; (middle and lower panels) immunoblot profiles with anti-Dl. In the middle panels, the amounts of sample applied for SDS-PAGE were optimized for quantification of the D1 protein band of 32 kDa, while in the lower panels 15-25 times as much sample was applied for the detection of the D1 protein fragments. Arrowheads indicate the position of the 32-kDa D1 protein band. with gentle stirring, with white light from a projector lamp, which was passed through two layers of heat-reflecting filters and a 10-cm-thick layer of water. The temperature was maintained at 20 OC with circulatingwater under thermostatic control. The light intensity was reduced by neutral density filters. When indicated, 20 mM DBMIB was added to give a final concentration of 0.2 mM just before illumination. Immediately after illumination, the suspension was divided into aliquots, frozen in liquid nitrogen, and stored at -80 "C. Polypeptide and Immunoblot Analyses. Samples were solubilized in 50 mM NaZCO3,50 mM dithiothreitol,2% SDS, and 12% (w/v) sucrose at 25 OC for 30 min and subjected to SDS-PAGE (Miyao & Murata, 1984) using a gel containing 6.5 M urea. The polyacrylamide concentration in the separation gel was 13%. Each lane of the gel was loaded with the equal amount of sample on the Chl basis: in the case of the PSII complexes, the sample amounts loaded were 0.6 pg and 5-100 ng of Chl for Coomassie staining and immunoblotting, respectively. After electrophoresis, the gel was stained with Coomassie brilliant blue R-250 or subjected to immunoblotting as follows. Separated polypeptides were electroblotted onto a nitrocellulose membrane (0.2 pm, Schleicher & Schuell) in the presenceof 0.05%SDS according to Towbin et al. (1979) with a semi-dry-type blot apparatus. After being blotted, the nitrocellulose membrane was blocked with 3% (w/v) gelatin and probed with 1000-fold-diluted antiserum raised in rabbit agairist spinach D1 protein (antiD1, a generous gift from Dr. M. Ikeuchi). Immunoreacted bands were further immunodecorated with goat antibodies against rabbit IgG conjugated with alkaline phosphatase (Jackson ImmunoResearch) and visualized by reaction with

nitroblue tetrazolium and bromochloroindolyl phosphate (BioRad) according to the manufacturer's instructions. The intensities of bands were quantified in terms of peak areas in densitograms recorded with a TLC scanner (CS-930, Shimadzu). The deviation of results was around 5%. Enzymes and Chemicals. Catalase from bovine liver and SOD from bovine erythrocytes were purchased from Sigma. Active oxygen scavengers and protease inhibitors were all purchased from Sigma, apart from aprotinin and SBTI, which were from Boehringer. All the protease inhibitors except for TPCK effectively inhibited the ability of tyrpsin to hydrolyze a syntheticsubstrate,Na-benzoyl-~,~-arginine-4-nitroanilide, as assayed by the method of Erlanger et al. (1961).

RESULTS Figure 1 shows the changes in polypeptide and immunoblot profiles for the three different PSII subcomplexes during photoinhibitory illumination. As seen from the Coomassiestained gels (upper panels), not only the D1 protein but also other intrinsic proteins were damaged in all the subcomplexes. When the PSII complexes were illuminated at pH 6.0 without electron acceptors (acceptor-side photoinhibition), Coomassiestained bands of all intrinsic proteins became smeared, and their positions on the gel were shifted closer to the origin with increasing duration of illumination: bands of the D1 and D2 proteins were affected first, and then those of the 43- and 47-kDa proteins of the core antenna and the a! subunit of Cyt b559 followed. Concomitantly, high-molecular-mass aggregates, seen as a smearing in the upper part of the gel, increased with illumination. By contrast, the stained bands of the three

Biochemistry, Vol. 33, No. 32, 1994 9725

D1 Protein Degradation and Active Oxygen extrinsicproteins of 33,23, and 16 kDa were not substantially affected. The smearing and mobility shift of protein bands and aggregate formation were most marked in the RC complexes illuminated at pH 8.0 in the presence of DBMIB (donor-side photoinhibition), and no distinct bands were observed after 40-min illumination. Similar changes in the polypeptide profiles were observed when the intensity of photoinhibitory light was reduced by 75% or when detergent was omitted from theillumination mixture,though the changes occurred more slowly (data not shown). When proteins were analyzed by immunoblottingusing antiD1 (Figure 1, middle and lower panels), it was clear that the D1 protein was degraded in almost the same way as previously reported for isolated PSII subcomplexes: we observed the gradual disappearance and mobility shift of the 32-kDa D1 protein band (Shipton & Barber, 1992; De Las Rivas et al., 1992,1993); the appearance of fragments of 23 and 16 kDa in the case of acceptor-sidephotoinhibition (Salter et al., 1992; De Las Rivas et al., 1993);and the appearance of three major fragments of about 24, 16, and 10 kDa in the case of donorside photoinhibition (De Las Rivas et al., 1992; Shipton & Barber, 1992). Immunoblots also revealed the appearance of a distinct band of 41 kDa at the onset of illumination, which has been reported to be a covalent adduct of the D1 protein and the a subunit of Cyt b559 (Barbato et al., 1992b), as well as the accumulationof high-molecular-massaggregatesduring prolonged illumination(e.g., De Las Rivas et al., 1992).Thus, it was evident that the illumination conditions in this study were similar to those in previous studies in terms of damage to the D1 protein. As judged from the densitograms of the immunoblots, the disappearance of the 32-kDa D1 protein band under illumination resulted mainly from the formation of high-molecular-mass aggregates. Decreased intensity of staining with Coomassie blue and decreased mobility of protein bands in SDS-PAGE are phenomena that are usually observed when proteins are exposed to active oxygen species (Davies, 1987). To examine the possible involvement of active oxygen in protein damage during illumination, the effects of active oxygen scavengers were investigated. In these experiments, the duration of illumination was fixed such that the amount of 23-24-kDa fragment, the putative primary degradation product of the D1 protein, was still increasing. The scavengers used were catalase (for H202), SOD (for 02-), histidine and DABCO (for l 0 2 ; Foote, 1976), and n-propyl gallate (for *OH and alkoxy1 radicals; Bors et al., 1989). In the PSII complexes subjected to acceptor-side photoinhibition, all six scavengers moreor less suppressedthe protein damage (Figure 2A). As shown by the Coomassie-stained gel, the extent of smearing and shifting of protein bands was reduced by the scavengers, most markedly by catalase, histidine, and n-propyl gallate. As seen in the immunoblot profiles, these scavengers also suppressed the decrease in the 32-kDa D1 protein band. Among the scavengers, n-propyl gallate was the most effective. Quantification from densitograms indicated that the 32-kDa band decreased to 47% of the darkcontrolvalue upon 30-min illumination in the absence of scavengers, while more than 80% remained after illumination in the presence of n-propyl gallate (Table 1). This value corresponds to an approximately 70% inhibition. Formation of the 23-kDa fragment was also suppressed by catalase, histidine, and n-propyl gallate but not by SOD or DABCO. In the RC complexes subjected to donor-side photoinhibition, in which the damage to intrinsic proteins was most marked, histidine and n-propyl gallate greatly suppressed the band

A

Illuminated

(a)

Illumi. D 1 2 3

Illumi. 0 1 4 5 6 '

Dl(32)w

- 23 -16

B

Illuminated 0 1 2 3 1 4 5 6 1 '

(b)

r 4

-10

FIGURE 2: Effects of catalase, SOD,and scavengers of active oxygen on protein damage by photoinhibitory illumination. The PSII complexes were suspended in Mes medium (pH 6.0) and the RC complexes in Tris medium (pH 8.0) as described in the legend to Figure 1. After a 5-min incubation in darkness at 20 OC, the suspension was supplemented with thedesignated additions,incubated for a further 5 min, and then illuminated with white light for 30 min in the presence or absence of 0.2 mM DBMIB. (A) PSII complexes illuminated at 8.0 mE/(m2-s) and pH 6.0. (B) RC complexes illuminated at 4.0 mE/(m2-s) and pH 8.0 in the presence of DBMIB. (a) Polypeptide profiles; (band c) immunoblot profiles with anti-Dl. D and I denote dark control and illuminated samples, respectively. (1) No additions; (2) 100 pg/mLcatalase; (3) 100 pg/mL SOD; (4) 10 mM histidine; (5) 1 mM DABCO; (6) 1 mM n-propyl gallate. Arrowheads indicate the position of the 32-kDa D1 protein band, and small arrows indicate those of bands of catalase and SOD.

smearing, the aggregation, and the formation of the 24-kDa fragment of the D1 protein (Figure 2B). A striking feature of Figure 2B is the damage to exogenously added proteins (catalase and SOD) under illumination. This was also the case when bovine serum albumin was present during illumination (not shown). This indicates that active oxygen species were even present in the surrounding medium when the RC complexes were illuminated. Table 1 shows the effects of active oxygen scavengers on the disappearance of the 32-kDa D1 protein and its fragmentation in the three typesof PSII subcomplex. With respect to the disappearance of the 32-kDa D1 protein, catalase had suppressive effects in the PSII and core complexes, but SOD was only effectivein the PSII complex. By contrast, histidine and n-propyl gallate showed remarkable suppression in all three subcomplexes. DABCO showed slight suppression in the PSII complex only. With respect to D1 protein fragmentation, the scavengers that suppressed the disappearance of the 32-kDa D1 protein were also effective in suppressing fragment formation in most cases. One exception was found with n-propyl gallate and the core complexes. n-Propyl gallate greatly suppressed the decrease in the 32-kDa band, but it increased the amount of 23-kDa fragment, even upon illumination for a shorter time (15 min, data not shown).

9126 Biochemistry, Vol. 33, No. 32, 1994

Miyao

Table 1: Effects of Catalase, SOD, and Scavengers of Active Oxygen Species on Degradation of the D1 Protein by Photoinhibitory Illumination of PSII Subcomplexes0 relative amounts of D1 protein and its fragments catalase His n-propy1 SOD illumination conditions no addition (0.1 mg/mL) (0.1 mg/mL) (10mM) gallate (1 mM) 1.31 1.10 1.77 PSII complex pH 6.0.30 min 32-kDa 1.00 (47) 1.27 1.43 1.02 0.53 1.oo 0.37 23-kDa 1.OO (4.0) 0.82 1.18 1.01 1.28 1.oo core complex pH 6.0,30 min 32-kDa 1.OO (40) 1.37 1.07 1.19 23-kDa 1.oo (9.3) 0.99 0.58 0.83 1.05 0.9 1 1.20 1.05 RC complex pH 6.3,4 min 32-kDa 1.00 (81) 1.10 1.01 1.05 0.78 23-kDa 1.oo (3.4) 1.03 0.67 RC complex pH 8.0, +DBMIB, 30 min 32-kDa 1.oo (47) 0.9 1 0.99 1.20 0.8 1 1.20 0.55 0.84 0.9 1 24-kDa 1.OO (1 3.8) 0.83 0.60 PSI1 subcomplexes were illuminated with white light in the presence of the designated additions at 20 OC. The experimental conditions were the same as those in the legends to Figures 1and 2. 32-kDa, 23-kDa, and 24-kDa stand for the intact D1 protein of 32 kDa, the 23-kDa N-terminal fragment, and the 24-kDa C-terminal fragment, respectively, and their amounts were quantified from densitograms of immunoblots with anti-Dl Values in parentheses represent percentages when the amount of intact D1 Drotein in the dark control samDle was taken as 100%. tYPe of PSII subcomplex

.

A

(a)

(b)

0 102030 0

0 1020 30 0

(c) 0 102030 0 min

t

B

(a) Illuminated

(b)

Illuminated

D l 2 3 4 5 6 0

D 1 2 " 3 " 4 5- 6 1 D